GuShiio.com Wen| vb arterial network
DeepSeek’s popularity is still continuing, and the deep reasoning model and deep thinking model fully demonstrate the advantages of a longer thought chain. Through dialogue with DeepSeek, users can get more comprehensive, clear, and high-quality answers.
Due to the need to interact frequently with C-end users, Internet medical platforms have used AI Q & A products in large quantities. In order to improve the accuracy of the answer content, many companies such as Chunyu Doctor, Jingdong Health, Ping An Health, Tencent Health, Weimai, and Zhiyun Health have connected DeepSeek’s large model capabilities to their products.
Internet medical and health service products or platforms that have been connected to DeepSeek, source: corporate public information
In fact, whether the platform is connected to DeepSeek or not, AI has penetrated more and more deeply into Internet medical services. However, the innovative thinking presented by DeepSeek has allowed the industry to see another new possibility.
Some inspirations from DeepSeek
The emergence of DeepSeek not only brings technological upgrades to the industry, but also brings inspiration for product and model innovation.
First of all, AI cannot replace doctors, but it can help create digital clones for doctors and replicate some of their service capabilities.
The shortage of doctor resources is a long-term problem, whether online or offline, but the many characteristics of Internet medical care make this phenomenon more obvious.
Although doctors can provide services on Internet medical platforms through multi-point practice identities, their labor relations still exist in the main offline practice institution, and practicing on third-party platforms mainly uses fragmented time. In the past, in order to fully activate debris resources, Internet medical platforms not only attracted more doctors to settle in, but also used order-based consultation products to mobilize doctors ‘time to meet patients’ consultation needs in a timely manner. This is mainly aimed at grassroots doctors or other doctors with relatively spare time.
For famous doctors and experts who have little fragment time and difficulty in seeing a doctor, digital avatar can play a more practical role. Digital doppelganger can respond to user-initiated consultation needs at any time based on fully and deeply learning expert knowledge, personalized diagnosis and treatment styles, etc.
Secondly, AI can make online medical services both rational and warm.
“The phenomenon of queuing for one hour and seeing a doctor for three minutes has been criticized in the traditional medical service system for many years. On the one hand, the fundamental reason is that doctors see too many patients and have no time to communicate too much with patients.
After Internet medical services have improved the efficiency of medical treatment, there is still a situation where doctors can finish speaking in a few sentences. This involves another important reason: not every doctor is good at expression, and not everyone can logically and clearly describe the causes and consequences of the diagnosis to patients.
When DeepSeek answers user questions in a deep thinking mode, it will give a thinking process and a systematic process. This is an idea that online medical services can learn from, helping doctors and other roles answer user questions more logically.
At the same time, the platform builds an interactive and evolvable digital image. Through the deep integration of data ethics, interactive design, and service models, it embeds a deep understanding of human dignity, emotional needs, and cultural differences into the technical logic, and keenly captures users ’emotions. and respond. This may also greatly improve users ‘impression of some doctors’ bad attitudes.“”
Finally, through analysis and in-depth reasoning of massive data, we can accurately gain insight into what unmet needs users have, thereby promoting business transformation.
In a comprehensive Internet medical platform, in addition to consultation and consultation, it will also generate data on various aspects such as search browsing, health monitoring, and medication habits. It is difficult to grasp the real needs of users by relying solely on data analysis in a single dimension or partial dimensions. The insight into real needs is based on the construction of multi-dimensional user portraits and highly relies on semantic reasoning of unstructured text; at the same time, it also needs to be based on key data Changes, track changes in user needs.
It can be seen that some Internet medical platforms are already iterating AI product and service capabilities along the above path.
Agents, digital people, innovative online consultation
Online consultation is the most representative service of Internet medical care. In order to alleviate the current shortage of doctor resources, the Internet medical platform promotes the intelligence of online consultations through the creation of carriers such as agents and digital people to replicate the ability of doctors and other service roles.
Starting in 2024, Alipay will begin to build intelligent agents for doctors. Zhang Tao, chief physician of Peking University People’s Hospital, is the first doctor to open an agent assistant on Alipay. The agent focuses on doctors ‘practice needs and provides AI tools to establish a digital avatar of doctors and amplify the service effectiveness of famous doctors.
Recently, Jingdong Health’s AI doctor agent products completed internal testing, and the first batch of AI digital avatars of doctor experts were launched in Jingdong Internet Hospital. AI digital avatar can deeply learn the doctor’s professional knowledge, way of thinking and expression habits, and answer patient consultations and questions 7*24 hours a day; it can also register patients, make appointments for the doctor’s own time, or provide post-diagnosis related services when needed.
Dr. Chunyu’s specialist digital doctor outputs service content based on clinical data that have been desensitized by top real doctors in various departments, combined with a massive medical knowledge base.
Not only doctors, but other roles in medical and health services are also beginning to be given agent functions.
Relying on the Beijing Medical Corporation’s Qianxun medical model, Jingdong Health has created a series of professional services for digital people. In addition to AI doctors, it also includes AI pharmacists, AI nutritionists, AI psychological counselors, etc.
While building doctor agents, Alipay also cooperates with health departments, medical insurance departments, medical institutions, specialties, etc. to create corresponding agents. For example, Zhejiang Health Commission and Alipay jointly built the first medical service agent Anjian Children. In December 2024, Anjian Children was upgraded to version 2.0 to provide users with famous doctor agents, intelligent report interpretation, intelligent health files, and intelligent medical treatment. Services such as accompanying medical services.
More roles and institutional agents further enrich online service resources and deepen the combination of online and offline services. From online consultation and consultation to services provided by agents and digital people, AI has brought many changes to Internet medical care.
In the past, online consultation included various forms such as text, voice, video, and telephone. Each form had its advantages and disadvantages. Users can choose according to their own circumstances. It is still the main form of communication between doctors and patients. Agents integrate multiple interaction forms and can give full play to the advantages of various forms.
Agents also realize the iteration from passive question answering to active question asking. In the traditional consultation process, when a user initiates a consultation request, he needs to ask a core question and a description of the core question. The platform will also give targeted prompts to allow the patient to supplement the medical condition description or medical record data, and then the doctor will make Answer and diagnosis. When agents communicate with patients, they can intervene more proactively, especially to conduct in-depth reasoning and simulate doctors ‘questioning logic.
At the same time, the agent uses the image of a human being to close the psychological distance with the user. At present, digital clones usually appear as the image of a doctor or other character, and collect the person’s voice, tone, expression habits and other characteristics, striving to achieve a state closer to the person when communicating with users; institutional agents exist as a virtual digital person with affinity and can be diverse. No matter which image it is, it can create a more realistic emotional connection with users.
Not just questions and answers, but also services
Writing copywriting, making plans, and making forms In many areas, DeepSeek’s in-depth thinking answers can inspire users and help them improve efficiency. Medical and health services are unique. Although Internet medical platforms can provide users with health suggestions, report interpretations, or nutrition plans, these contents are essentially information and it is difficult to fully solve users ‘actual problems. Therefore, session-as-a-service has become an exploration idea to solve this situation.
Arterial Network learned that Micromedicine Holdings is currently developing new AI products for C-end users, which are already in the internal testing stage and will soon be launched; the product will integrate multiple advanced large model capabilities, strengthen the micromedicine medical large model capabilities, and can provide users with safe, compliant, effective, and more personalized services.
Since Jingdong Health’s AI Health Intelligent Body Kangkang was launched in July 2024, as of now, relying on the medical health e-commerce and the Internet medical health service ecosystem, it has integrated more than 52,000 medical institution registration resources and more than 190 health services, not only Including Jingdong’s own service matrices such as purchasing medicines, Jingdong Internet Hospital, Jingdong Home Express Check-in, and Jingdong Nurse Arriving Home, there are also more third-party health services connected to it.
In September 2024, Alipay released AI Health Butler. Provide users with AI health services such as finding a doctor, reading reports, accompanying doctors, asking for medical insurance, and managing health. For example, while users ask for medical insurance, they can directly handle medical insurance business using the AI-recommended portal when needed.
Behind the service is also Alipay’s deep connection and integration of internal and external resources, including the various agents mentioned above. At the same time, Alipay also announced that it will open agent collaboration to the medical and pan-health industries to attract more cooperation resources.
In February 2025, CareAI, an intelligent health management application for micro-pulse, was connected to DeepSeek V3 and R1 model capabilities and used it in multiple scenarios such as pre-consultation, doctor recommendation, service recommendation, and case management.
In order to support the above services, CareAI not only relies on Weimai’s self-operated service resources, but also connects online and offline medical and health service resources of Weimai’s contracted cooperative hospitals, as well as third-party ecological cooperative companies such as insurance, accompanying medical services, door-to-door care, family health care, remote monitoring and other resources.
Tencent Health recently connected to the DeepSeek model through Tencent Cloud to form a dual-mode combination with Tencent’s Mixed Model. At the same time, Tencent Health is accelerating the creation of a large-scale medical service application platform to provide large-scale medical services and intelligent applications to more than 1000 hospitals across the country; landing hospitals can choose to use DeepSeek or Tencent Hunyuan to provide various medical services and health knowledge through AI assistants, and develop more humane medical treatment plans and health management for patients.
In short, in addition to outputting information, AI needs to solve problems for users through services as accurately as possible. Solving user problems through accurate recommendations is also one of the commercialization paths of the platform.
On the one hand, over time, the Internet medical platform carries more and more service content. In reality, not all users have the professional ability and patience to find and search for the services they need. As mentioned earlier, AI can conduct in-depth reasoning based on multi-dimensional user data to capture user needs, and even understand their needs more clearly than the user himself; products or services recommended to users based on deep insight into user needs can only form a higher transformation.
On the other hand, the Internet medical platform plays an important role in the out-of-hospital market and is a link connecting medicine, nursing care, medicine, insurance, etc. The platform will explore personalized user service models with these participants. In this process, improving the matching between resources and users, allowing medical, nursing, medicine, and insurance to efficiently find target users, and allowing users to accurately obtain medical, nursing, medicine, and insurance are the value points of AI.
On the new path, we still need to pay attention to AI application boundaries
In general, the in-depth reasoning represented by DeepSeek is not only a technical tool, but also points out a new path for Internet medical care: upgrading from simple online consultation to smart medical brains, providing each individual user with all the medical resources they need. Promote personalized full life cycle health services to become a reality. Judging from the current product status of major platforms, there is still a gap in achieving this goal. However, a critical step from zero to one has been taken.
Across the industry, DeepSeek is enthusiastic about transforming its medical service model. However, the more this is the case, the more we need to remain calm. For Internet medical care, the bottom line of AI applications still needs to be adhered to.
Previously, the “Internet Diagnosis and Treatment Supervision Rules (Trial)” has clearly stipulated that other personnel, artificial intelligence software, etc. are not allowed to falsely use or replace the doctor himself to provide diagnosis and treatment services; it is strictly prohibited to use artificial intelligence, etc. to automatically generate prescriptions.
In real conditions, the behavior of AI prescribing prescriptions is relatively easier to define and avoid, while the red line of AI replacing doctors in providing diagnosis and treatment services requires vigilance. In services provided by doctor agents and digital avatars, there may be blurred boundaries between health advice and diagnosis and treatment behaviors, especially when it comes to disease diagnosis and medication advice. The industry needs to remain in awe at all times and innovate within the framework of laws, regulations and ethics to go further.
